Hindawi Publishing Corporation Leukemia Research and Treatment Volume 2012, Article ID 179402, 14 pages doi:10.1155/2012/179402

Review Article Important in the Pathogenesis of 5q- Syndrome and Their Connection with Ribosomal Stress and the Innate Immune System Pathway

Ota Fuchs1, 2

1 Institute of Hematology and Blood Transfusion, U Nemocnice 1, 128 20 Prague 2, Czech Republic 2 Center of Experimental Hematology, First Medical Faculty, Charles University, Institute of Pathological Physiology, 128 53 Prague 2, Czech Republic

Correspondence should be addressed to Ota Fuchs, [email protected]

Received 25 September 2011; Revised 6 November 2011; Accepted 14 November 2011

Academic Editor: Daniela Cilloni

Copyright © 2012 Ota Fuchs. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Myelodysplastic syndrome (MDS) with interstitial deletion of a segment of the long arm of 5q [del(5q)] is characterized by bone marrow erythroid hyperplasia, atypical megakaryocytes, thrombocythemia, refractory anemia, and low risk of progression to acute myeloid leukemia (AML) compared with other types of MDS. The long arm of contains two distinct commonly deleted regions (CDRs). The more distal CDR lies in 5q33.1 and contains 40 -coding genes and genes coding microRNAs (miR-143, miR-145). In 5q-syndrome one allele is deleted that accounts for haploinsufficiency of these genes. The mechanism of erythroid failure appears to involve the decreased expression of the ribosomal protein S14 (RPS14) and the upregulation of the p53 pathway by ribosomal stress. Friend leukemia virus integration 1 (Fli1) is one of the target genes of miR145. Increased Fli1 expression enables effective megakaryopoiesis in 5q-syndrome.

1. Introduction putative tumor suppressor and a transcriptional activator of interferon and interferon-stimulated genes. IRF1 dosage ex- Approximately 15% of patients with MDS have abnormali- periments demonstrated that 2 patients with 5q-syndrome ties of chromosome 5 [1]. These abnormalities include inter- retained both copies of this gene [12]. Thus, IRF1 maps out- stitial deletion of a segment of the long arm of chromosome side the common deleted segment of the 5q-chromosome, 5q [del(5q), 5q-syndrome], monosomy, and unbalanced and the same result was obtained in the case of EGR1 (epi- translocations. 5q-syndrome as MDS category was defined dermal growth receptor 1) [13, 14]. by the World Health Organization (WHO) [2],anditischar- Molecular mapping techniques defined the region of acterized by refractory macrocytic anemia with dyserythro- gene loss in two patients with the 5q-syndrome and unchar- poiesis, transfusion dependence, normal to elevated platel- acteristically small 5q deletions (5q31–q33) [14]. The allelic et counts, hypolobated and nonlobated megakaryocytes, loss of 10 genes localized to 5q23-qter [centromere-CSF2 female preponderance, a favourable prognosis, and low risk (colony-stimulating factor 2/granulocyte-macrophage/)- of progression to AML compared with other types of MDS EGR1 (early growth response1)-FGFA (fibroblast growth [3–10]. Many research groups analysed chromosome 5q factor acidic)-GRL (glucocorticoid receptor)-ADRB2 (β-2 deletions in patients with 5q-syndrome. We will shortly adrenergic receptor)-CSF1R (colony-stimulating factor 1)- describe these studies from historical point of view and not SPARC (secreted protein, acidic, cysteine-rich/osteonectin/)- for relevance in the pathogenesis. GLUH1(human glutamate receptor 1)-NKSF1 (NK-cell- Deletion of interferon regulatory factor-1 gene (IRF1) stimulating factor chain 1)-FLT4 (Fms-related tyrosine mapped to chromosome 5q31 was detected [11]. IRF1 is a kinase 4)-telomere] was investigated in peripheral blood cell 2 Leukemia Research and Treatment fractions. Gene dosage experiments demonstrated that 5q23.2 Cytokine cluster CSF2, EGR1, NKSF1, and FLT4 were retained on the 5q23.3 (IL-3, IL-4, IL-5, IL-13, GM-CSF) 5q-chromosome in both patients and that FGFA was 5q31.1 PP2A 5q31.2 MDS/AML retained in one patient, thus placing these genes outside the (CTNNA1, EGR1, CDC25C) CDR1 critical region. GRL, ADRB2, CSF1R, SPARC,andGLUH1 5q31.3 were shown to be deleted in both patients. The proximal 5q32 5q33.1 5q-syndrome CDR2 breakpoint is localized between EGR1 and FGFA in one 5q33.2 (RP514, SPARC) patient and between FGFA and ADRB2 in the other, and the 5q33.3 5q34 distal breakpoint is localized between GLUH1 and NKSF1 NPMI 5q35.1 in both patients [14]. Pulsed-field gel electrophoresis was 5q35.2 used to map the 5q deletion breakpoints, and breakpoint- 5q35.3 specific fragments were detected with FGFA probe in the granulocyte but not the lymphocyte fraction of one patient. Figure 1: Schematic diagram of human chromosome 5q map This study has established the critical region of gene loss of showing commonly deleted regions. the 5q-chromosome in the 5q-syndrome, giving the location for a putative tumor-suppressor gene in the 5.6 Mb region between FGFA and NKSF1 [14]. Boultwood et al. [15] characterised the commonly delet- mDia1 acts as a node in a tumor-suppressor network that in- ed region (CDR) in a study involving sixteen 5q-syndrome volves multiple 5q gene products. patients to a 1.5 Mb interval located at 5q32-5q33 between D5S413 marker and GLRA1 (glycine receptor subunit α-1). This region contains PDE6A (phosphodiesterase 6A), CSF1R, 2. The Possible Role of Candidate Genes CD74 (CD74/cluster of differentiation 74/molecule, major from CDR in 5q-Syndrome histocompatibility complex, class II invariant chain), TCOF1 (Treacher-Collins-Franceschetti syndrome 1), ANXA6 (an- All the 40 genes within the CDR were sequenced, and nexin A6), SPARC,andFAT2 (FATtummorsuppressorhom- no mutations were found [21]. This finding suggests that olog 2, also known as cadherin family member 8 or multiple 5q-syndrome may be a “one-hit” cancer in contrast to the epidermal growth factor-like domains protein 1) genes. “two-hit hypothesis” for the mechanism of cancer develop- Advanced MDS and AML had a large del(5q) which con- ment described by Knudson [22]. Knudson’s two-hit model tained a distinct 5q31 CDR [16, 17]. of tumour suppressor genes supposes that two mutations are It was shown that MDS arises from the transformation required to cause a tumour, one occurring in each of the two of a multipotent hematopoietic stem cell (HSC) or myeloid- alleles of the gene. Many reports described candidate tumour committed progenitor cell [18, 19]. These data suggest that suppressors that do not conform to this standard definition, the gene or genes that are inactivated in the 5q-syndrome will including haploinsufficient genes requiring inactivation of be expressed in normal hematopoietic stem and progenitor only one allele and genes inactivated not by mutation but cells. The expression of all 40 genes assigned to the CDR by rather epigenetic hypermethylation [23]. Thus, a dosage ef- means of the Ensembl program was thus examined in normal fect as the result of the loss of a single allele of a gene (hap- human bone marrow CD34+ cells by means of RT-PCR. loinsufficiency) may be responsible for the 5q-syndrome. Of the 40 genes in the CDR, 33 were expressed in CD34+ Haploinsufficiency of multiple genes mapping to the CDR at cells and, therefore, represented candidate genes since they 5q31-32 may contribute to the pathogenesis of 5q-syndrome were expressed within the HSC/progenitor cell compartment [24–27]. Haploinsufficiency may be a driving force in cancer [15]. [28]. Cytogenetic mapping of commonly deleted regions Boultwood et al. [21] compared the transcriptome of the (CDRs) centered on 5q31 and 5q32-5q33 identified can- CD34+ cells in a group of 10 patients with the 5q-syndrome didate tumor-suppressor genes, including the transcription using the Affymetrix Gene Chip U133 Plus 2.0 array platform factor Egr1/Krox20 and the cytoskeletal remodeling protein, with the transcriptome of CD34+ cells from 16 healthy alpha-catenin on 5q31, and the ribosomal subunit protein control subjects and 14 patients with refractory anemia and a RPS14 on 5q32-5q33 (Figure 1). Although each acts as a normal karyotype. The majority of the genes assigned to the tumor suppressor, alone or in combination, no molecular CDR of the 5q-syndrome at 5q31-q32 showed a reduction in mechanism accounts for how defects in individual 5q candi- expression levels in patients with the 5q-syndrome, consis- dates may act as a lesion driving MDS or contributing to tent with the loss of one allele. Candidate genes showing malignant progression in MPN (myeloproliferative neo- haploinsufficiency in the 5q-syndrome included the tumour plasms). One candidate gene that resides between the con- suppressor gene SPARC and RPS14, a component of the 40S ventional del(5q)/5q-MDS-associated CDRs is DIAPH1 ribosomal subunit. Two genes mapping to the CDR, RBM22 (5q31.3). DIAPH1 encodes the mammalian diaphanous- (RNA-binding motif 22) and CSNK1A1 (casein kinase 1, related formin, mDia1. mDia1 has critical roles in actin α1), showed more than 50% reduction in gene expression, remodeling in cell division and in response to adhesive and consistent with the downregulation of the remaining allele. migratory stimuli. Eisenmann et al. [20] examined evidence, RBM22 plays a role in splicing and nuclear translocation of with a focus on mouse gene-targeting experiments, that the calcium binding protein ALG2 (apoptosis-linked gene 2) Leukemia Research and Treatment 3 and causes deregulation of apoptosis [29]. CSNK1A1 regu- erythroid lineage. DBA is characterized by a moderate to lates the Hedgehog signal pathway which governs cell growth severe anemia with normal neutrophil and platelet counts in cooperation with Wnt signaling [30, 31]. This study and a marked reduction in number of red cell precursors in identified several significantly deregulated gene pathways in an otherwise normocellular bone marrow. Several other patients with the 5q-syndrome, and gene pathway analysis bone marrow failure syndromes are also associated with de- data supported the proposal that SPARC might play a role in fects in ribosome biogenesis and function [25, 38]. the pathogenesis of the 5q-syndrome. A murine model of the human 5q-syndrome was devel- oped in which haploinsufficiency of the Cd74-Nid67 (CD74 antigen gene—nerve growth factor-induced differentiation 2.1. Role of SPARC in 5q-Syndrome. Lehmann et al. [32]fur- clone 67 gene) region containing the RPS14 gene on mouse ther studied the hematopoietic system in SPARC-null mice. chromosome 18 maps CDR of the human 5q-syndrome and These mice showed significantly lower platelet counts com- contains 8 known genes [49]. This was achieved using Cre- pared to wild-type animals. Although hemoglobin, hemat- loxP recombination to delete this region. An Lmo2-Cre trans- ocrit, and mean corpuscular volume (MCV) were lower in gene was used to restrict the deletion to the hematopoietic mice lacking SPARC, differences were not statistically sig- compartment. Two genes (Ndst1/glucosaminyl N-deacet- nificant. SPARC-null mice showed a significantly impaired ylase/N-sulfotransferase 1 gene/and Cd74) from this segment ability to form erythroid burst-forming units (BFU-E). How- on mouse chromosome 18 have been excluded as candidate ever, no significant differences were found in the forma- genes. RPS14 was the major candidate gene in relation to tion of erythroid colony-forming units (CFU-E), granulo- the phenotype [49]. Mice with one deleted allele displayed cyte/monocyte colony-forming units (CFU-GM), or megak- hypocellularbonemarrowasaconsequenceofadefectin aryocyte colony-forming units (CFU-Mk) in these animals. hematopoietic progenitor function that could be rescued by These authors concluded that many of the genes within the p53 inactivation. This murine model also displayed a macro- CDR associated with the 5q-syndrome exhibit significantly cytic anemia with dysplastic features in the bone marrow decreased expression and that SPARC, as a potential tumor and monolobulated megakaryocytes, common features of suppressor gene, may play some but not the key role in the 5q-syndrome. These changes correlated with an increase in pathogenesis of this disease. a population of highly p53 positive cells in the bone marrow andwithelevatedapoptosis(seeSection 4). 2.2. RPS14 and Ribosomopathies. Ebertetal.[33] used an Pellagatti et al. [50] investigated the expression profiles RNA-mediated interference- (RNAi-) based approach to dis- of a large group of ribosomal- and translation-related genes covery of the 5q- disease gene. They found that partial loss in the CD34+ cells isolated from bone marrow samples of of function of the ribosomal subunit protein RPS14 phe- 15 MDS patients with 5q-syndrome, 18 MDS patients with nocopies the disease in normal haematopoietic progenitor refractory anemia and a normal karyotype, and 17 healthy cells. The forced expression of RPS14 with a lentiviral cDNA controls. U133 Plus 2.0 arrays (Affymetrix) expression vector rescued the disease phenotype in patient- covering over 47 000 transcripts representing 39 000 human derived bone marrow cells. In addition, they identified a genes were used in this study. The expression data for selected block in the processing of preribosomal RNA in RPS14- genes were validated using real-time quantitative PCR with deficient cells. This block was functionally equivalent to predeveloped TaqMan assays (Applied Biosystems). The the defect in Diamond-Blackfan anemia, linking the molec- expression profiles of 579 probe sets for genes coding riboso- ular pathophysiology of the 5q-syndrome to a congenital mal (229 for large 80S ribosomal subunit/RPL/and syndrome causing bone marrow failure [34]. These results 176 for small 40S ribosomal subunit/RPS/) and for genes indicate that the 5q-syndrome is caused by a defect in ribo- coding translation-related factors (149 for eukaryotic somal protein function and suggest that RNAi screening is translation initiation factors/EIF/and 25 for eukaryotic an effective strategy for identifying causal haploinsufficiency translation elongation factors/EEF/) were analysed [50]. 55 disease genes. Multiple different RPS14 shRNAs decreased genes were differentially expressed. 49 from these 55 genes the ratio of erythroid to megakaryocytic cells produced in (89%) showed lower expression in the 5q-syndrome patient vitro. These RPS14 shRNAs decreased also the ratio of group in comparison with MDS patients with refractory erythroid to myeloid cells and caused the apoptosis of anemia and a normal karyotype and in comparison with differentiating erythroid cells. The decreased level of RPS14 healthy controls. These data were compared with data about gene expression in 5q-syndrome is not caused by the aberrant the defective expression of genes for ribosomal proteins and methylation of the RPS14 gene promoter [35]. The 5q-syn- translation-related factors in DBA published by Pellagatti et drome belongs to ribosomopathies that are disorders in al. [50]. Three genes (RPL28, RPS14, and EEF1D) expression which genetic abnormalities cause impaired ribosome bio- was found downregulated in both diseases (MDS 5q- genesis and function [25]. syndrome and DBA) [50, 51]. In addition, the expression of Inherited mutations for several ribosomal protein genes two pro-3poptotic genes, TNFRSF10B (tumor necrosis factor (RPS19, RPS27A, RPS26, RPS24, RPS17, RPS15, RPS10, receptor superfamily, member 10b gene) and BAX (BCL2- RPS7, RPL5, RPL11, RPL35a,andRPL36) cause Diamond- associated X protein gene), was upregulated in both diseases. Blackfan anemia (DBA, OMIM#205900) [34, 36–46]. Pellagatti et al. [50] suggested that the deregulation observed Josephs [47] and Diamond and Blackfan [48] reported the in ribosomal gene expression and translation-related gene first cases of this rare disease almost exclusively affecting the expression in the CD34+ cells isolated from bone marrow 4 Leukemia Research and Treatment

adjacent to the distal boundary of the CDR and is also deleted in most patients with 5q-syndrome. Starczynowski et al. [59] 4oS 5oS Normal identified Toll-interleukin-1 receptor domain-containing p53 MDM2 erythrocyte adaptor protein (TIRAP) and tumor necrosis factor receptor- associated factor-6 (TRAF6) as respective targets of these miRNAs. These miRNAs target corresponding genes through  Normal cell partial base pairing to the 3 -UTR of the target genes. These targets were aberrantly upregulated in 5q-syndrome Apoptosis [59]. Downregulation of the miRNAs (miR-145 and miR- 146a) led to an upregulation of IL-6 that was dependent on TRAF6. Elevated IL-6 was found also in patients with 5q- syndrome [60]. TIRAP is known to lie upstream of TRAF6 in p53 activation 4oS excess iron innate immune signaling. Knockdown of miR-145 and miR- 5oS in erythrocyte 146a together or enforced expression of TRAF6 in mouse p53 RPL5, 11 and 23 MDM2 HSPCs resulted in thrombocytosis, mild neutropenia, and RPS27L megakaryocytic dysplasia. A subset of mice transplanted with TRAF6-expressing marrow progressed either to marrow Ribosomal haploinsufficiency failure or acute myeloid leukemia. Thus, inappropriate activation of innate immune signals in HSPCs phenocopies several clinical features of 5q-syndrome. Starczynowski et al. have recently showed that loss of two miRNAs, miR-145 and Oxidative stress miR-146a, results in leukemia in a mouse model [61]. and hemolysis Patients with 5q-syndrome have decreased expression of Figure 2: Ribosomal stress in 5q-syndrome. On the top is schemat- miR-145 and increased expression of Fli-1 [62]. Overex- + ically shown normal erythroid cell with normal ribosome biogene- pression of miR-145 or inhibition of Fli-1 in CD34 cells sis and p53 levels. On the bottom is erythroid cell where p53 is acti- decreases megakaryocyte production, while inhibition of vated by ribosomal stress secondary to RPS14 haploinsufficiency. miR-145 or overexpression of Fli-1 has the reciprocal effect. Ribosomal proteins RPL5, 11, 23, and RPS27L bind HDM2 and These findings have been validated in vivo using transgenic activate p53. The ribosomal protein S27-like (RPS27L) is induced mice. Moreover, the combined loss of miR-145 and RPS14 by p53-activating signals and promotes apoptosis [53, 54]. To date, cooperates to alter erythroid-megakaryocytic differentiation the role of oxidative stress in MDS has not been fully elucidated in a manner similar to the 5q-syndrome. Taken together, [55–57]. these findings demonstrated for the first time that coordinate deletion of a microRNA and a protein-coding gene (Figure 3) contributes to the phenotype of the 5q-syndrome [62]. Other research groups found normal expression levels of samples of MDS patients with 5q-syndrome are secondary + to RPS14 haploinsufficiency. These data and similar data of miR-145 in most CD34 cells isolated from bone marrow Sridhar et al. [52] support the hypothesis that the 5q-syn- samples of MDS patients with 5q-syndrome [63, 64]. drome belongs to ribosomopathies, disorders of impaired Remarkably, the guardians of the genome p53, p73, and p63 ribosomal biogenesis (Figure 2). play a role in control of most of the known tumor suppressor miRNAs [65, 66]. Thus, it is possible that one allele of miR- 145 is lost in 5q-syndrome but the second allele of miR-145 3. The Role of miR-145 and miR-146a in the is overexpressed after p53 or related genes activation. A puta- 5q-Syndrome tive tumor suppressing miRNA, miR-145, has been shown to be decreased in various human cancers [67, 68], and it In mammalian cells, miRNAs (microRNAs) are the most decreases the apoptosis and proliferation rate of colorectal abundant family of small noncoding RNAs that regulate cancer cells and B-cell lymphoma cell lines [69, 70]. mRNA translation through the RNA interference pathway Zhang et al. [71] demonstrated that miR-145 targets a or target-specific mRNA for degradation [58]. In general, it putative binding site in the 3-UTR of the Friend leukemia appears that the major function of miRNAs is in develop- virus integration 1 (Fli-1) gene and that miR-145 abundance ment, differentiation, and homoeostasis, which is indicated is inversely related to Fli-1 expression in colon cancer tissues. by studies showing aberrant miRNA expression during the Some other targets of miR-145 are important regulators of development of cancer. cell apoptosis and proliferation, such as c-Myc and IRS-1 Starczynowski et al. [59] examined expression of microR- [62, 72]. IRS-1, a docking protein for both the type 1 insulin- NAs (miRNAs) encoded on chromosome 5q as a possible like growth factor receptor and the insulin receptor, delivers cause of haploinsufficiency. They showed that deletion of antiapoptotic and antidifferentiation signals. MiR-145 also chromosome 5q correlates with loss of two miRNAs that are downregulates the protooncogene c-Myc, whose aberrant abundant in hematopoietic stem/progenitor cells (HSPCs), expression is associated with aggressive and poorly differ- miR-145, and miR-146a. The miR-145 gene maps within the entiated tumors. Zhang et al. [73] demonstrated that DNA CDR of the 5q-syndrome [21], and the miR-146a gene lies fragmentation factor 45 (DFF45) expression is controlled at Leukemia Research and Treatment 5 the translational level by miR-145, using bioinformatic and synthesis. Initiation factor eIF4E binds the 5 cap struc- proteomic techniques. DFF45 is caspase-3 or caspase-7 ture present on all cellular mRNAs. Its ability to associ- substrate that must be cleaved before apoptotic DNA frag- ate with initiation factors eIF4G and eIF4A, forming the mentation can proceed [74, 75]. eIF4F complex, brings the mRNA to the 43S complex dur- Changes in the expression of miR-146a have been impli- ing the initiation of translation. Binding of eIF4E to eIF4G cated in both the development of multiple cancers and in the is inhibited in a competitive manner by 4E-BP1. Phospho- negative regulation of inflammation induced via the innate rylation of 4E-BP1 decreases the affinity of this protein for immune response. Furthermore, miR-146a expression is eIF4E, thus favouring the binding of eIF4G and enhancing driven by the transcription factor NF-κB(nuclearfactorkap- translation. The truncated protein 4E-BP1 is almost com- paB), which has been implicated as an important causal link pletely unphosphorylated and exerts a long-term inhibitory between inflammation and carcinogenesis. Williams et al. effect on the availability of eIF4E, thus contributing to the [76]andLietal.[77] reviewed the evidence for a role of inhibition of protein synthesis and the growth-inhibitory miR-146a in innate immunity and cancer and assessed and proapoptotic effects of p53. whether changes in miR-146a might link these two biological The murine double minute 2 protein (MDM2) and hu- responses. man analogue (HDM2) function as a link between ribosome The role of miR-146a in hematopoiesis was investigated biogenesis and the p53 pathway. MDM2 or HDM2 proteins by using retroviral infection and overexpression of miR-146a function in two major ways: (1) as an E3 ubiquitin ligase that in mouse hematopoietic stem/progenitor cells, followed by recognizes the N-terminal transactivation domain (TAD) bone marrow transplantations [78]. miR-146a is mainly ex- of the p53 tumor suppressor and (2) as an inhibitor of pressed in primitive hematopoietic stem cells and T lympho- p53 transcriptional activation [88–90]. Having a short half- cytes. Overexpression of miR-146a in hematopoietic stem life, p53 is normally maintained at low levels in unstressed cells, followed by bone marrow transplantation, resulted mammalian cells by continuous ubiquitination and subse- in a transient myeloid expansion, decreased erythropoiesis, quent degradation by the 26S proteasome. This is primarily and impaired lymphopoiesis in select anatomical locations. due to the interaction of p53 with the RING-finger ubiq- Enforced expression of miR-146a also impaired bone mar- uitin E3 ligases, MDM2 or HDM2. MDM2 interacts with row reconstitution in recipient mice and reduced survival a variety of ribosomal proteins, including RPL5, RPL11, of hematopoietic stem cells. These results indicate that RPL23, RPL26, and RPS7 [91–99]. These interactions, which miR-146a, a lipopolysaccharide- (LPS-) induced miRNA, typically involve the acidic domain and sometimes the regulates multiple aspects of hematopoietic differentiation adjacent zinc finger of MDM2, interfere with the inhibitory and survival. Furthermore, the consequences of miR-146a functions of this region of MDM2 and contribute to p53 expression in hematopoietic cells mimic some of the re- activation (Figure 2). As first exemplified for RPL11 [92], ported effects with acute LPS exposure. these interactions increase when ribosome biogenesis is dis- Third gene for miRNA in the CDR is miR-143 [62]. rupted, in a situation termed “ribosomal biogenesis stress” Cordes et al. [79] and Boettger et al. [80] proposed that miR- or “nucleolar stress” [100, 101]. Mechanistically, ribosomal 145 and miR-143 regulate smooth muscle cells (SMC) fate stress causes translocation of free ribosomal proteins from and plasticity and play the important role in cardiovascular the nucleolus to the nucleoplasm [94, 102], where they bind development. MDS patients with 5q deletion were charac- MDM2 [94]. The increased binding of ribosomal proteins terized by decreased levels of miR-143 and miR-378 mapped to MDM2 augments cellular p53 activity, leading to growth within the commonly deleted region at 5q32 [81]. arrest and coupling deficient protein synthesis with cessation of cell proliferation. 4. The Importance of p53 in the Molecular However, it is not clear whether ribosomal protein regu- Mechanism of 5q-Syndrome lation of MDM2 is specific to some, but not all ribosomal proteins. Sun et al. [103] showed that RPL29 and RPL30, Dysregulation of ribosome biogenesis through insufficiency two ribosomal proteins from the 60S ribosomal subunit, do of ribosomal subunits has been demonstrated to activate a not bind to MDM2 and do not inhibit MDM2-mediated p53 p53 checkpoint and regulate developmental pathways [25, suppression, indicating that the ribosomal protein regulation 82–86]. The p53 pathway provides a surveillance mechanism of the MDM2-p53 feedback loop is specific. for translation in protein synthesis and for genome integrity. The transcription factor p53 is encoded by TP53 tumor sup- 4.2. Activation of p53 and Upregulation of the p53 Pathway pressor gene. Target genes of p53 regulate critical processes, in 5q-Syndrome. Results of several laboratories described in such as apoptosis, cell cycle arrest, DNA repair, senescence, Section 4.1 showed that upregulation of the p53 pathway is and cellular metabolism. a common response to haploinsufficiency of ribosomal pro- teins. Pellagatti et al. [104] carried out the analysis of the p53 4.1. Activation of p53 by Ribosome Dysfunction in Model Sys- pathway in the 5q-syndrome. They used Affymetrix arrays tems. Activation of p53 stimulates proteasome-dependent [50]. Ten genes in the p53 pathway (FAS/tumor necrosis fac- truncation of eIF-4E-binding protein 1 (4E-BP1) [87]. The tor receptor superfamily, member 6 gene/,CD82/cluster of translational inhibitor protein 4E-BP1 regulates the availabil- differentiation 82 gene/, WIG1/wild type p53-induced gene ity of polypeptide chain initiation factor eIF4E for protein 1gene/,CASP3/cysteine-aspartic acid protease, caspase 3 6 Leukemia Research and Treatment

Initiating mutation of cell growth and proliferation, it is considered as a com- plementary opposite factor to the oncoprotein c-Myc. The oncoprotein c-Myc can counteract the p53’s action to a great 5q deletion degree. The cell has developed sophisticated mechanisms during evolution to balance the effect of both these factors. It has been known for a long time that there is a negative correlation between p53 and c-Myc. Although previous miR145, miR146a RPS14 report [106] suggested that this might involve transcrip- haploinsufficiency haploinsufficiency tional repression the precise mechanism of p53-mediated repression of c-Myc is not fully understood. Sachdeva et al. p53 activation [107] have recently demonstrated that p53 induces miR-145 and subsequently represses c-Myc. Thus, identification of Target gene(s) regulation HDM2 (Fli 1 upregulation miR145 miR-145, that was described in Section 3 as a new player in cMyc upregulation) this p53 regulatory network (Figure 3), provides at least one p53 target genes ff activation aspect of how the cell balances the e ects of p53 and c-Myc [108].

Thrombocytosis Erythroid defect 4.4. TP53 Mutation in 5q-Syndrome. TP53 mutations were often detected in high-risk MDS and AML with del(5q) 5q syndrome [110, 111]. Jadersten¨ et al. [112] described a patient with 5q-syndrome with del(17p) and TP53 mutation at disease Adjunct cytogenetic progression. They demonstrated the TP53 mutant clone abnormalities already at the initial diagnosis. Consequently, they analysed 55 patients with low- and int-1 risk MDS and del(5q) and AML identified 10 patients (18%) with TP53 mutations [113]. Figure 3: Schematic diagram of the development of the 5q-syn- Moreover, they found a correlation between TP53 mutation drome. The roles of the haploinsufficiency of RPS14, miR145, and and strong nuclear p53 protein expression in bone marrow miR146 and possible progression to AML by adjunct cytogenetic progenitors [113]. Patients with mutations had signifi- abnormalities are shown. Fli1 upregulation stimulates transcrip- cantly worse outcome. Clonal heterogeneity in low-risk 5q- tionally HDM2 [109] and inhibits partially p53 activation caused ffi syndrome patients may be of importance for the prognosis by ribosomal stress resulting from RPS14 haploinsu ciency. and therapy [113].

5. Role of Further Genes in 5q-Syndrome gene/, SESN3/sestrin 3 gene/, TNFRSF10B, MDM4/murine 5.1. The Role of Other Genes Positioned on 5q Chromosome. double minute 4 gene/, BAX, DDB2/damage-specific DNA ffi binding protein 2 gene/, and BID2/BH3-interacting domain The murine model for the 5q-syndrome with haploinsu - death agonist 2 gene/) were significantly deregulated. All ciency of the Cd74-Nid67 region described in the Section 2.2 these genes with exception of MDM4 (negative regulator of showed that other gene in this region may play an additional p53, structurally related to MDM2 [105]) were expressed role. Dctn4 gene encodes Dynactin 4, a subunit of the dyn- at higher levels in CD34+ cells from 5q-syndrome patients actin complex. The dynactin complex binds cargo, such as vesicles and organelles, to cytoplasmic dynein for retrograde in comparison with healthy controls. MDM4 is expressed ffi at lower level in 5q-syndrome patients. Further upregulated microtubule-mediated tra cking [114]. Retroviral expres- genes in 5q-syndrome (RPS27L/ribosomal protein S27-like sion in megakaryocytes of dynamitin (p50), which disrupts gene/,PHLDA3/pleckstrin homology-like domain, family A, dynactin-dynein function, inhibits proplatelet elongation member 3 gene/, BCL11B/B-cell lymphoma/leukemia 11B [115]. Patel et al. [115] concluded that while continuous gene/, and FDXR/ferredoxin reductase gene/) are p53 targets polymerization of microtubules is necessary to support the [21, 50]. Strong p53 expression was confirmed not only enlarging proplatelet mass, the sliding of overlapping micro- on the mRNA level but also on the protein level by an tubules is a vital component of proplatelet elongation. Thus, immunohistochemical analysis. Activation of p53 as a result the inhibition of proplatelet elongation might reduce platelet of ribosomal stress seems to play an important role in the formation and lead to thrombocytopenia [116]. The further 5q-syndrome. Upregulation of the p53 pathway offers a new gene Rbm22 was described in Section 2. RBM22 belongs to the SLT11 family; it has been reported to be involved potential therapeutic target in the human 5q-syndrome. This 2+ therapy is possible only if the critical tumor suppressor in prespliceosome assembly and to interact with the Ca - function of p53 will be preserved. signaling protein ALG-2 [117]. RBM22 is underexpressed in 5q-syndrome and deregulates probably apoptosis [29]. In addition to the minimal CDR on chromosome 5q33.1 4.3. The Relation between p53 and miR145. The tumor sup- in 5q-syndrome, a second, proximal CDR on chromosome pressor p53 is a master gene regulator capable of regulating 5q31.2 contains genes EGR1, CTNNA1 (gene-encoding al- numerous genes. Since p53 plays a critical role in suppression pha-catenin), and HSPA9 (gene-encoding mortalin, GRP75, Leukemia Research and Treatment 7

PBP74, and MTHSP75). Loss of this CDR contributes to a biogenesis [137]. One of the consequences of miR145 and more aggressive MDS or AML phenotype [16, 17]. miR146a depletion or TIRAP and TRAF6 activation in the The gene for nucleophosmin (NPM1)localizedonchro- innate immune system pathway is overexpression of IL-6 mosome 5q35.1 (Figure 1) is deleted in many cases of which activates Fli-1 gene expression [59, 78, 138–140]. 5q- MDS. Targeted knockout of one allele of NPM1 leads EKLF (erythroid Kruppel-like˝ factor, also named KLF1) to genetic instability [118–120]. NPM1 heterozygous mice is a zinc finger transcription factor that plays a prominent develop both myeloid and lymphoid malignancies. The role during erythroid development [141–143]. Human gene bone marrow of these mice is hypercellular with evidence EKLF is positioned on chromosome 19p13.12-13 [144]. of dysplasia of the erythroid and megakaryocytic lineages. Commitment towards megakaryocyte versus erythroid Immature erythroid cells accumulate in this bone marrow. blood cell lineages occurs in the megakaryocyte-erythroid The NPM1 haploinsufficiency in patients with large 5q progenitor (MEP). EKLF restricts megakaryocytic differen- deletions contributes to the phenotype of MDS. Moreover, tiation to the benefit of erythrocytic differentiation, and this NPM1 is the most commonly mutated gene in patients with effect is mediated by the inhibition of Fli-1 recruitment to normal karyotype AML [121–123]. megakaryocytic and Fli-1 genes promoters [130, 145–147]. A marked increase in the number of circulating platelets in EKLF null mouse was found [148]. 5.2. The Role of Genes Not Positioned on Chromosome 5. We studied the role of the levels of mRNA for transcriptions factors Fli-1 (Friend leukemia virus integration 1) and 6. Conclusions and Perspectives EKLF (erythroid Kruppel-like˝ factor, also named KLF1) in mononuclear cells isolated from bone marrow and periph- Candidate genes were studied using gene deletions or similar eral blood of MDS patients with 5q-syndrome in comparison approaches to determine whether they have a role in 5q- with patients with low risk MDS without 5q chromosome syndrome, a subtype of MDS. A major advance occurred in abnormality and with healthy controls [124, 125]. We found 2008 when RNA interference-based approach for all 40 genes increased Fli1 mRNA levels and decreased EKLF mRNA located within the 5q- CDR was used to discover the 5q- levels in 5q-syndrome. The decreased expression of miR- disease gene. A defect in the function of a ribosomal protein 145 might be responsible for increased Fli-1 levels in 5q- RPS14 that is important for the proper processing of the 40S syndrome (see Section 3). We suggest that increased Fli-1 ribosomal subunit has been implicated in the pathophysiol- expression in mononuclear cells of MDS patients with 5q- ogy of the 5q-syndrome. The RPS14 gene is located in the syndrome is further stimulated by the increased Fli-1 level deleted region, and it seems that this loss affects erythroid and by the decreased EKLF level. differentiation. Haploinsufficiency of RPS14 gene in 5q- Fli-1 is a member of the Ets family of transcription fac- syndrome is associated with deregulation of ribosomal- and tors. Fli-1 was originally identified as a protooncogene in translation-related genes. Dutt et al. [149] found that p53 Friend virus-induced erythroleukemia in mice [126]. The accumulates selectively in the erythroid lineage in primary founding member of this family, the oncogene v-ets, was human hematopoietic progenitor cells following expression discovered in the genome of avian leukosis virus E26. Ets of shRNAs targeting RPS14 or RPS19. Thus, 5q-syndrome is transcription factors bind to DNA elements containing the similarly as Diamond Blackfan anemia, where RPS 19 is the consensus sequence GGA(A/T). Fli-1 is preferentially ex- most commonly mutated gene, a ribosomopathy [25, 34, 36, pressed in hematopoietic cells and in vascular endothelial 150, 151]. cells. Pulse-field gel analysis has localized the Fli-1gene Induction of p53 in both diseases, but not in control sam- within 240 kb of the Ets-1 locus on mouse chromosome 9 ples, led to lineage-specific accumulation of p21WAF1,CIP1, and on human chromosome 11q23, suggesting that these two also known as cyclin-dependent kinase inhibitor 1 or CDK- ets genes arose by gene duplication from a common ancestral interacting protein 1 (a proliferation arrest determinating gene [127, 128]. Human Fli-1 contains nine exons which protein that in humans is encoded by the CDKN1A extend over approximately 120 kb [129]. gene located on chromosome 6p21.2). The induction of Fli-1 plays an important role during the normal develop- p21WAF1,CIP1 caused cell cycle arrest in erythroid progenitor ment of megakaryocytes. Fli-1 knockout mice produce small, cells. Pharmacological inhibition of p53 rescued the ery- undifferentiated megakaryocytic progenitors with abnormal throid defect, while a compound that activates p53 through features. Fli-1 represses the transcription of EKLF gene [130] inhibition of HDM2, nutlin-3, selectively impaired erythro- and activates the promoters of several megakaryocyte-spe- poiesis [149]. The increased apoptosis was observed in the cific genes (glycoprotein IX and thrombopoietin receptor 5q- mouse bone marrow [35]. In response to diverse stresses, genes promoters) [131, 132]. Fli-1 gene promoter is upreg- the tumor suppressor p53 differently regulates its target ulated by Fli-1 and other Ets factors (Ets1, Ets2, and Elf1) genes, inducing cell cycle arrest, apoptosis, or senescence. but is inhibited by Tel [133]. Transcription factor PU.1 (also The p53 is critical regulator of hematopoietic stem cell (HSC) known as Spi-1) is also a positive regulator of the Fli1 gene behavior and plays an important role in regulating HSC [134]. On the other hand, PU.1 inhibits transcription factor quiescence, self-renewal, apoptosis, and aging. The p53 GATA-1 function and erythroid differentiation by blocking promotes HSC quiescence [152, 153]. The mutant p53 GATA-1 DNA binding [135, 136]. Both, PU.1 and Fli-1 protein can provide escape from apoptosis and facilitates also directly activate common target genes involved in ribosome the entry of HSC to the cell cycle. Mutations in TP53 are 8 Leukemia Research and Treatment relatively uncommon in cases of primary (de novo) MDS but complex and includes inhibition of a wide array of proin- have high incidence in patients with therapy-related MDS flammatory cytokines, such as interleukin-6 (IL-6), and [154–156]. Garderet et al. [157] found defective erythroid upregulation of T-helper-secreted cytokines, such as IL-2 proliferation but not differentiation capacity related to and IFN-gamma [180]. It has been also described that IL- RPS14 gene haploinsufficiency. They used in vitro model of 2 subsequently activates natural killer (NK) cells. In view of erythropoiesis [158, 159] in which mature red blood cells the recently revealed role of IL-6 in the pathogenesis of the are generated from human progenitor cells to analyze cell 5q-syndrome [59, 138], it is tempting to speculate that the proliferation and differentiation in a homogeneous erythroid responses induced by lenalidomide therapy in patients with population with RPS14 gene haploinsufficiency in culture. this disorder are due to downregulation of IL-6. Therefore, The enucleation capacity of 5q- clones remained unchanged. IL-6 expression and/or upstream regulators, such as TRAF6, Therefore, the decreased erythroid maturation and subse- may represent targets, which suggests that agents capable of quent anemia seen in 5q-syndrome are unlikely caused by a inhibiting the TRAF6-IL-6 axis may be clinically useful for specific blockade of late differentiation and are probably the the management of 5q-syndrome [59, 138]. consequences of the proliferative defect of both pathological  Lenalidomide inhibits the malignant clone and upregu- and residual nondeleted clones in the patient sbonemarrow lates the SPARC and RPS14 genes expression [181, 182]. Both [160]. these genes are localized in CDR on chromosome 5. Induc- Treatment of zebrafish models of ribosomopathies with tion of miR-143 and miR-145 in CD34+ cells of MDS patients L-leucine results in an improvement of anemia and develop- with the del(5q) abnormality after exposure to lenalidomide mental defects in both diseases (Diamond Blackfan anemia correlates with clinical response [179]. The beneficial effect and 5q-syndrome) [161]. Cmejlova et al. [162] demonstrated of lenalidomide in patients with 5q-syndrome is associated that the efficiency of mRNA translation is significantly with significant increases in the proportion of bone marrow depressed in cells derived from Diamond Blackfan anemia erythroid precursor cells and in the frequency of clonogenic patients, consistent with a pathogenic ribosomal defect. progenitor cells, a substantial improvement in the hemato- L-leucine is an essential branched chain amino acid that is poiesis-supporting potential of bone marrow stroma and sig- known to modulate protein synthesis by enhancing trans- nificant alterations in the adhesion profile of bone mar- lation [162–166]. L-leucine has been used to treat Dia- row CD34+ cells [183]. Although lenalidomide efficiently mond Blackfan anemia patients [162, 167]. Treatment of reduced a larger fraction of the del(5q) stem cells, some RPS19,andRPS14-deficient zebrafish embryos and human rare and phenotypically distinct quiescent del(5q) stem cells hematopoietic progenitor CD34+ cells with L-leucine result- remained resistant to lenalidomide [112, 184]. Over time, ed in partial reversal of both the anemia and the develop- lenalidomide resistance developed in most of the patients mental defects and provided evidence for a common sig- with recurrence or expansion of the del(5q) clone and clinical naling pathway involving the mTOR (mammalian target of and cytogenetic progression. The resistant clone, insensi- rapamycin) integrating growth-promoting and stress signals tive to lenalidomide, overexpresses p53 and sequencing con- [161]. firmed TP53 mutation [185]. The presence of easily detect- From a translational point of view, the identification of able subclones with inactivated p53, and thereby a more the genes and miRNAs involved in the pathogenesis of the malignant potential, is important and has prognostic value. 5q-syndrome may support the development of novel tar- geted therapeutic strategies. The immunomodulatory drug lenalidomide (CC-5013; REVLIMID, Celgene Corporation), a structural analogue of thalidomide, is indicated for the Acknowledgments treatment of the 5q-syndrome MDS patients, rendering 67% This work was supported by the research intention VZ of patients transfusion independent and inducing cytoge- 00023736 from the Ministry of Health of the Czech Republic, netic responses in over 40% of them [168–173]. In contrast, Grant MSM 0021620808 and Grant LC 06044 from Ministry in a large multicenter trial involving transfusion: dependent of Education, Youth and Sport of the Czech Republic. MDS patients without del(5q) only 26% achieved transfu- sion independence with infrequent cytogenetic improvement [174]. The selective and specific activity of lenalidomide in MDS remains undefined. Gene expression profiling and References disease- and therapy-associated proteome changes in the sera of MDS patients were used for monitoring and predicting [1] D. Haase, U. Germing, J. Schanz et al., “New insights into the the response to therapy [175, 176]. Wei et al. [177]provided prognostic impact of the karyotype in MDS and correlation with subtypes: evidence from a core dataset of 2124 patients,” evidence that lenalidomide is selectively cytotoxic to del(5q) Blood, vol. 110, no. 13, pp. 4385–4395, 2007. cells as a result of inhibition of the haplodeficient dual [2] R. P. Hasserjian, M. M. Le Beau, A. F. List et al., WHO Classi- specificity phosphatases, Cdc25C and PP2A. Phosphoryla- fication of Tumors of Haematopoietic and Lymphoid Tissues, tion of MDM2 and Fli-1 is important for the level and the International Agency for Research on Cancer Press, Lyon, activity of both these proteins involved in the process of France, 2007. erythroid failure in 5q-syndrome, and the level of PP2A [3] H. Van Den Berghe, J. J. Cassiman, and G. David, “Distinct regulates accumulation and degradation of p53 and Fli-1 haematological disorder with deletion of long arm of No. 5 [178, 179]. The mechanism of action of lenalidomide is chromosome,” Nature, vol. 251, no. 5474, pp. 437–438, 1974. Leukemia Research and Treatment 9

[4] H. Van Den Berghe, K. Vermaelen, and C. Mecucci, “The 5q- [21] J. Boultwood, A. Pellagatti, H. Cattan et al., “Gene expression anomaly,” Cancer Genetics and Cytogenetics, vol. 17, no. 3, pp. profiling of CD34+ cells in patients with the 5q- syndrome,” 189–255, 1985. British Journal of Haematology, vol. 139, no. 4, pp. 578–589, [5]S.D.NimerandD.W.Golde,“The5q-abnormality,”Blood, 2007. vol. 70, no. 6, pp. 1705–1712, 1987. [22] A. G. Knudson, “Mutation and cancer: statistical study [6] P. Mathew, A. Tefferi, G. W. Dewald et al., “The 5q- syn- of retinoblastoma,” Proceedings of the National Academy of drome: a single-institution study of 43 consecutive patients,” Sciences of the United States of America, vol. 68, no. 4, pp. 820– Blood, vol. 81, no. 4, pp. 1040–1045, 1993. 823, 1971. [7] J. Boultwood, S. Lewis, and J. S. Wainscoat, “The 5q- [23] A. J. W. Paige, “Redefining tumour suppressor genes: excep- syndrome,” Blood, vol. 84, no. 10, pp. 3253–3260, 1994. tions to the two-hit hypothesis,” Cellular and Molecular Life [8] H. Van Den Berghe and L. Michaux, “5q-, twenty-five years Sciences, vol. 60, no. 10, pp. 2147–2163, 2003. later: a synopsis,” Cancer Genetics and Cytogenetics, vol. 94, [24] B. L. Ebert, “Deletion 5q in myelodysplastic syndrome: a no. 1, pp. 1–7, 1997. paradigm for the study of hemizygous deletions in cancer,” [9] A. A. N. Giagounidis, U. Germing, J. S. Wainscoat, J. Boult- Leukemia, vol. 23, no. 7, pp. 1252–1256, 2009. wood, and C. Aul, “The 5q-syndrome,” Hematology, vol. 9, [25] A. Narla and B. L. Ebert, “Ribosomopathies: human disor- no. 4, pp. 271–277, 2004. ders of ribosome dysfunction,” Blood, vol. 115, no. 16, pp. [10] A. Mohamedali and G. J. Mufti, “Van-den Berghe’s 5q- 3196–3205, 2010. syndrome in 2008,” British Journal of Haematology, vol. 144, [26] M. Tormo, I. Marugan,´ and M. Calabuig, “Myelodysplastic no. 2, pp. 157–168, 2009. syndromes: an update on molecular pathology,” Clinical and [11] C. L. Willman, C. E. Sever, M. G. Pallavicini et al., “Deletion Translational Oncology, vol. 12, no. 10, pp. 652–661, 2010. of IRF-1, mapping to chromosome 5q31.1, in human leuke- [27] M. S. Davids and D. P. Steensma, “The molecular patho- mia and preleukemic myelodysplasia,” Science, vol. 259, no. genesis of myelodysplastic syndromes,” Cancer Biology and 5097, pp. 968–971, 1993. Therapy, vol. 10, no. 4, pp. 309–319, 2010. [12] J. Boultwood, C. Fidler, S. Lewis et al., “Allelic loss of IRF1 [28] A. H. Berger and P. P. Pandolfi, “Haplo-insufficiency: a in myelodysplasia and acute myeloid leukemia: retention of driving force in cancer,” Journal of Pathology, vol. 223, no. 2, IRF1 on the 5q- chromosome in some patients with the 5q- pp. 137–146, 2011. syndrome,” Blood, vol. 82, no. 9, pp. 2611–2616, 1993. [29] P. Montaville, Y. Dai, C. Y. Cheung et al., “Nuclear translo- [13] M. M. Le Beau, R. Espinosa, W. L. Neuman et al., “Cytoge- cation of the calcium-binding protein ALG-2 induced by the netic and molecular delineation of the smallest commonly RNA-binding protein RBM22,” Biochimica et Biophysica Acta, deleted region of chromosome 5 in malignant myeloid vol. 1763, no. 11, pp. 1335–1343, 2006. diseases,” Proceedings of the National Academy of Sciences of [30] J. Jia, C. Tong, B. Wang, L. Luo, and J. Jiang, “Hedgehog the United States of America, vol. 90, no. 12, pp. 5484–5488, signalling activity of smoothened requires phosphorylation 1993. by protein kinase A and casein kinase I,” Nature, vol. 432, no. [14] J. Boultwood, C. Fidler, S. Lewis et al., “Molecular mapping 7020, pp. 1045–1050, 2004. of uncharacteristically small 5q deletions in two patients with the 5q- syndrome: delineation of the critical region on 5q and [31] A. Hammerlein,¨ J. Weiske, and O. Huber, “A second protein kinase CK1-mediated step negatively regulates Wnt signal- identification of a 5q- breakpoint,” Genomics,vol.19,no.3, β pp. 425–432, 1994. ling by disrupting the lymphocyte enhancer factor-1/ - catenin complex,” Cellular and Molecular Life Sciences, vol. [15] J. Boultwood, C. Fidler, A. J. Strickson et al., “Narrowing and 62, no. 5, pp. 606–618, 2005. genomic annotation of the commonly deleted region of the 5q- syndrome,” Blood, vol. 99, no. 12, pp. 4638–4641, 2002. [32] S. Lehmann, J. O’Kelly, S. Raynaud, S. E. Funk, E. H. Sage, ffl [16] R. J. Jaju, J. Boultwood, F. J. Oliveret et al., “Molecular and H. P. Koe er, “Common deleted genes in the 5q- cytogenc definition of the critical deleted region in the 5q- syndrome: thrombocytopenia and reduced erythroid colony syndrome,” Genes Cancer, vol. 22, pp. 251–256, formation in SPARC null mice,” Leukemia,vol.21,no.9,pp. 1998. 1931–1936, 2007. [17] N. Zhao, A. Stoffel, P. W. Wang et al., “Molecular delineation [33] B. L. Ebert, J. Pretz, J. Bosco et al., “Identification of RPS14 of the smallest commonly deleted region of chromosome 5 as a 5q- syndrome gene by RNA interference screen,” Nature, in malignant myeloid diseases to 1-1.5 Mb and preparation vol. 451, no. 7176, pp. 335–339, 2008. of a PAC-based physical map,” Proceedings of the National [34] S. R. Ellis and P. E. Gleizes, “Diamond Blackfan anemia: ri- Academy of Sciences of the United States of America, vol. 94, bosomal proteins going rogue,” Seminars in Hematology, vol. no. 13, pp. 6948–6953, 1997. 48, pp. 89–96, 2011. [18] M. L. Heaney and D. W. Golde, “Myelodysplasia,” The New [35] A. Valencia, J. Cervera, E. Such, M. A. Sanz, and G. F. Sanz, England Journal of Medicine, vol. 340, no. 21, pp. 1649–1660, “Lack of RPS14 promoter aberrant methylation supports the 1999. haploinsufficiency model for the 5q- Syndrome,” Blood, vol. [19] L. Nilsson, I. Astrand-Grundstrom, I. Arvidsson et al., “Isola- 112, no. 3, p. 918, 2008. tion and characterization of hematopoietic progenitor/stem [36] N. Draptchinskaia, P. Gustavsson, B. Andersson et al., “The cells in 5q-deleted myelodysplastic syndromes: evidence for gene encoding ribosomal protein S19 is mutated in Dia- involvement at the hematopoietic stem cell level,” Blood, vol. mond-Blackfan anaemia,” Nature Genetics, vol. 21, no. 2, pp. 96, no. 6, pp. 2012–2021, 2000. 169–175, 1999. [20] K. M. Eisenmann, K. J. Dykema, S. F. Matheson et al., “5q- [37] H. T. Gazda, A. Grabowska, L. B. Merida-Long et al., “Ribos- myelodysplastic syndromes: chromosome 5q genes direct a omal protein S24 gene is mutated in Diamond-Blackfan tumor-suppression network sensing actin dynamics,” Onco- anemia,” American Journal of Human Genetics, vol. 79, no. gene, vol. 28, no. 39, pp. 3429–3441, 2009. 6, pp. 1110–1118, 2006. 10 Leukemia Research and Treatment

[38] K. A. Ganapathi and A. Shimamura, “Ribosomal dysfunction [55] M. J. Farquhar and D. T. Bowen, “Oxidative stress and and inherited marrow failure,” British Journal of Haematol- the myelodysplastic syndromes,” International Journal of ogy, vol. 141, no. 3, pp. 376–387, 2008. Hematology, vol. 77, no. 4, pp. 342–350, 2003. [39] M. F. Campagnoli, U. Ramenghi, M. Armiraglio et al., [56] H. Ghoti, J. Amer, A. Winder, E. Rachmilewitz, and E. Fibach, “RPS19 mutations in patients with Diamond-Blackfan ane- “Oxidative stress in red blood cells, platelets and polymor- mia,” Human Mutation, vol. 29, no. 7, pp. 911–920, 2008. phonuclear leukocytes from patients with myelodysplastic [40] R. Cmejla, J. Cmejlova, H. Handrkova et al., “Identification syndrome,” European Journal of Haematology, vol. 79, no. 6, of mutations in the ribosomal protein L5 (RPL5) and pp. 463–467, 2007. ribosomal protein L11 (RPL11) genes in Czech patients with [57] B. Novotna, Y. Bagryantseva, M. Siskova, and R. Neuwirtova, diamond-blackfan anemia,” Human Mutation,vol.30,no.3, “Oxidative DNA damage in bone marrow cells of patients pp. 321–327, 2009. with low-risk myelodysplastic syndrome,” Leukemia Re- [41] J. M. Lipton and S. R. Ellis, “Diamond-blackfan anemia: search, vol. 33, no. 2, pp. 340–343, 2009. diagnosis, treatment, and molecular pathogenesis,” Hematol- [58] D. P. Bartel, “MicroRNAs: target recognition and regulatory ogy/Oncology Clinics of North America, vol. 23, no. 2, pp. 261– functions,” Cell, vol. 136, no. 2, pp. 215–233, 2009. 282, 2009. [59] D. T. Starczynowski, F. Kuchenbauer, B. Argiropoulos et al., [42] L. Doherty, M. R. Sheen, A. Vlachos et al., “Ribosomal “Identification of miR-145 and miR-146a as mediators of the protein genes RPS10 and RPS26 are commonly mutated in 5q-syndrome phenotype,” Nature Medicine, vol. 16, no. 1, pp. diamond-blackfan anemia,” American Journal of Human 49–58, 2010. Genetics, vol. 86, no. 2, pp. 222–228, 2010. [60] E. Terpos, E. Verrou, A. Banti, V. Kaloutsi, A. Lazaridou, and [43] E. E. Devlin, L. DaCosta, N. Mohandas, G. Elliott, and ff D. M. Bodine, “A transgenic mouse model demonstrates a K. Zervas, “Bortezomib is an e ective agent for MDS/MPD dominant negative effect of a point mutation in the RPS19 syndrome with 5q- anomaly and thrombocytosis,” Leukemia gene associated with Diamond-Blackfan anemia,” Blood, vol. Research, vol. 31, no. 4, pp. 559–562, 2007. 116, no. 15, pp. 2826–2835, 2010. [61] D. T. Starczynowski, R. Morin, A. McPherson et al., “Ge- [44] J. Hoefele, A. M. Bertrand, M. Stehr et al., “Disorders of nome-wide identification of human microRNAs located in sex development and Diamond-Blackfan anemia: is there an leukemia-associated genomic alterations,” Blood, vol. 117, pp. association?” Pediatric Nephrology, vol. 25, no. 7, pp. 1255– 595–607, 2011. 1261, 2010. [62] M. Kumar, A. Narla, A. Nonami et al., “Coordinate 1oss [45] E. Ito, Y. Konno, T. Toki, and K. Terui, “Molecular pathogen- of a microRNA and protein-coding gene cooperate in the esis in Diamond-Blackfan anemia,” International Journal of pathogenesis of 5q- syndrome,” Blood, vol. 118, pp. 4663– Hematology, vol. 92, no. 3, pp. 413–418, 2010. 4673, 2011. [46] I. Boria, E. Garelli, H. T. Gazda et al., “The ribosomal basis of [63] J. Boultwood, A. Pellagatti, A. N. J. McKenzie, and J. S. Wain- diamond-blackfan anemia: mutation and database update,” scoat, “Advances in the 5q-syndrome,” Blood, vol. 116, no. 26, Human Mutation, vol. 31, no. 12, pp. 1269–1279, 2010. pp. 5803–5811, 2010. [47] H. W. Josephs, “Anemia of infancy and early childhood,” [64] H. Votavova, M. Grmanova, M. Dostalova Merkerova et al., Medicine, vol. 15, pp. 307–451, 1936. “Differential expression of microRNAs in CD34+ cells of 5q- [48] L. K. Diamond and K. D. Blackfan, “Hypoplastic anemia,” syndrome,” Journal of Hematology and Oncology, vol. 4, p. 1, Am.J.Dis.Child., vol. 56, pp. 464–467, 1938. 2011. [49] J. L. Barlow, L. F. Drynan, D. R. Hewett et al., “A p53- [65] H. I. Suzuki, K. Yamagata, K. Sugimoto, T. Iwamoto, S. Kato, dependent mechanism underlies macrocytic anemia in a and K. Miyazono, “Modulation of microRNA processing by mouse model of human 5q-syndrome,” Nature Medicine, vol. p53,” Nature, vol. 460, no. 7254, pp. 529–533, 2009. 16, no. 1, pp. 59–66, 2010. [66] L. Boominathan, “The guardians of the genome (p53, [50] A. Pellagatti, E. Hellstrom-Lindberg,¨ A. Giagounidis et al., TA-p73, and TA-p63) are regulators of tumor suppressor ffi “Haploinsu ciency of RPS14 in 5q- syndrome is associated miRNAs network,” Cancer and Metastasis Reviews, vol. 29, with deregulation of ribosomal- and translation-related no. 4, pp. 613–639, 2010. genes,” British Journal of Haematology, vol. 142, no. 1, pp. 57– [67] M. Ozen, C. J. Creighton, M. Ozdemir, and M. Ittmann, 64, 2008. “Widespread deregulation of microRNA expression in hu- [51] H. T. Gazda, A. T. Kho, D. Sanoudou et al., “Defective riboso- man prostate cancer,” Oncogene, vol. 27, no. 12, pp. 1788– mal protein gene expression alters transcription, translation, 1793, 2008. apoptosis, and oncogenic pathways in Diamond-Blackfan anemia,” Stem Cells, vol. 24, no. 9, pp. 2034–2044, 2006. [68]Y.WangandC.G.L.Lee,“MicroRNAandcancer—focuson apoptosis,” Journal of Cellular and Molecular Medicine, vol. [52] K. Sridhar, D. T. Ross, R. Tibshirani, A. J. Butte, and P. 13, no. 1, pp. 12–23, 2009. L. Greenberg, “Relationship of differential gene expression profiles in CD34+ myelodysplastic syndrome marrow cells to [69] Y. Akao, Y. Nakagawa, and T. Naoe, “MicroRNA-143 and - disease subtype and progression,” Blood, vol. 114, no. 23, pp. 145 in colon cancer,” DNA and Cell Biology,vol.26,no.5,pp. 4847–4858, 2009. 311–320, 2007. [53] H. He and Y. Sun, “Ribosomal protein S27L is a direct p53 [70] M. Sachdeva and Y. Y. Mo, “miR-145-mediated suppression target that regulates apoptosis,” Oncogene, vol. 26, no. 19, pp. of cell growth, invasion and metastasis,” American Journal of 2707–2716, 2007. Translational Research, vol. 2, no. 2, pp. 170–180, 2010. [54] J. Li, J. Tan, L. Zhuang et al., “Ribosomal protein S27-like, a [71] J. Zhang, H. Guo, H. Zhang et al., “Putative tumor suppressor p53-inducible modulator of cell fate in response to genotoxic miR-145 inhibits colon cancer cell growth by targeting stress,” Cancer Research, vol. 67, no. 23, pp. 11317–11326, oncogene friend leukemia virus integration 1 gene,” Cancer, 2007. vol. 117, no. 1, pp. 86–95, 2011. Leukemia Research and Treatment 11

[72] B. Shi, L. Sepp-Lorenzino, M. Prisco, P. Linsley, T. Deangelis, [88] J. Momand, H. H. Wu, and G. Dasgupta, “MDM2-master and R. Baserga, “Micro RNA 145 targets the insulin receptor regulator of the p53 tumor suppressor protein,” Gene, vol. substrate-1 and inhibits the growth of colon cancer cells,” 242, no. 1-2, pp. 15–29, 2000. Journal of Biological Chemistry, vol. 282, no. 45, pp. 32582– [89] S. Fang, J. P. Jensen, R. L. Ludwig, K. H. Vousden, and A. 32590, 2007. M. Weissman, “Mdm2 is a RING finger-dependent ubiquitin [73] J. Zhang, H. Guo, G. Qian et al., “MiR-145, a new regulator protein ligase for itself and p53,” Journal of Biological of the DNA Fragmentation Factor-45 (DFF45)-mediated Chemistry, vol. 275, no. 12, pp. 8945–8951, 2000. apoptotic network,” Molecular Cancer, vol. 9, article 211, [90] H. V. Clegg, K. Itahana, and Y. Zhang, “Unlocking the 2010. Mdm2-p53 loop: ubiquitin is the key,” Cell Cycle, vol. 7, no. [74] C. C. Chiu, C. H. M. Y. Lin, and K. Fang, “Etoposide (VP-16) 3, pp. 287–292, 2008. sensitizes p53-deficient human non-small cell lung cancer [91] M. S. Dai and H. Lu, “Inhibition of MDM2-mediated p53 cells to caspase-7-mediated apoptosis,” Apoptosis, vol. 10, no. ubiquitination and degradation by ribosomal protein L5,” 3, pp. 643–650, 2005. Journal of Biological Chemistry, vol. 279, no. 43, pp. 44475– [75] X. Liu, H. Zou, C. Slaughter, and X. Wang, “DFF, a 44482, 2004. heterodimeric protein that functions downstream of caspase- [92] M. A. E. Lohrum, R. L. Ludwig, M. H. G. Kubbutat, M. 3 to trigger DNA fragmentation during apoptosis,” Cell, vol. Hanlon, and K. H. Vousden, “Regulation of HDM2 activity 89, no. 2, pp. 175–184, 1997. by the ribosomal protein L11,” Cancer Cell,vol.3,no.6,pp. [76]A.E.Williams,M.M.Perry,S.A.Moschos,H.M.Larner- 577–587, 2003. Svensson, and M. A. Lindsay, “Role of miRNA-146a in the [93] Y. Zhang, G. W. Wolf, K. Bhat et al., “Ribosomal protein L11 regulation of the innate immune response and cancer,” Bio- negatively regulates oncoprotein MDM2 and mediates a p53- chemical Society Transactions, vol. 36, no. 6, pp. 1211–1215, dependent ribosomal-stress checkpoint pathway,” Molecular 2008. and Cellular Biology, vol. 23, no. 23, pp. 8902–8912, 2003. [77] L. Li, X.-P. Chen, and Y.-J. Li, “MicroRNA-146a and human [94] K. P. Bhat, K. Itahana, A. Jin, and Y. Zhang, “Essential role disease,” Scandinavian Journal of Immunology, vol. 71, pp. of ribosomal protein L11 in mediating growth inhibition- 227–231, 2010. induced p53 activation,” EMBO Journal, vol. 23, no. 12, pp. [78] D. T. Starczynowski, F. Kuchenbauer, and J. Wegrzyn, 2402–2412, 2004. “MicroRNA-146a disrupts hematopoietic differentiation and [95] M. S. Dai, S. X. Zeng, Y. Jin, X. X. Sun, L. David, and H. Lu, survival,” Experimental Hematology, vol. 39, pp. 167–178, “Ribosomal protein L23 activates p53 by inhibiting MDM2 2011. function in response to ribosomal perturbation but not to translation inhibition,” Molecular and Cellular Biology, vol. [79] K. R. Cordes, N. T. Sheehy, M. P. White et al., “MiR-145 24, no. 17, pp. 7654–7668, 2004. and miR-143 regulate smooth muscle cell fate and plasticity,” [96] Y. Ofir-Rosenfeld, K. Boggs, D. Michael, M. B. Kastan, Nature, vol. 460, no. 7256, pp. 705–710, 2009. and M. Oren, “Mdm2 regulates p53 mRNA translation [80] T. Boettger, N. Beetz, S. Kostin et al., “Acquisition of the through inhibitory interactions with ribosomal protein L26,” contractile phenotype by murine arterial smooth muscle cells Molecular Cell, vol. 32, no. 2, pp. 180–189, 2008. depends on the Mir143/145 gene cluster,” Journal of Clinical [97] D. Chen, Z. Zhang, M. Li et al., “Ribosomal protein S7 as a Investigation, vol. 119, no. 9, pp. 2634–2647, 2009. novel modulator of p53-MDM2 interaction: binding to [81] M. Dostalova Merkerova, Z. Krejcik, H. Votavova, M. Belick- MDM2, stabilization of p53 protein, and activation of p53 ova, A. Vasikova, and J. Cermak, “Distinctive microRNA function,” Oncogene, vol. 26, no. 35, pp. 5029–5037, 2007. expression profiles in CD34+ bone marrow cells from pa- [98] Y. Zhang and H. Lu, “Signaling to p53: ribosomal proteins tients with myelodysplastic syndrome,” European Journal of find their way,” Cancer Cell, vol. 16, no. 5, pp. 369–377, 2009. Human Genetics, vol. 19, pp. 313–319, 2011. [99] Y. Zhang, J. Wang, Y. Yuan et al., “Negative regulation of [82] L. Panic,´ J. Montagne, M. Cokaric,´ and S. Volarevic,´ “S6- HDM2 to attenuate p53 degradation by ribosomal protein ffi haploinsu ciency activates the p53 tumor suppressor,” Cell L26,” Nucleic Acids Research, vol. 38, no. 19, Article ID Cycle, vol. 6, no. 1, pp. 20–24, 2007. gkq536, pp. 6544–6554, 2010. [83] N. Danilova, K. M. Sakamoto, and S. Lin, “Ribosomal protein [100] D. G. Pestov, Z. Strezoska, and L. F. Lau, “Evidence of p53- S19 deficiency in zebrafish leads to developmental abnormal- dependent cross-talk between ribosome biogenesis and the ities and defective erythropoiesis through activation of p53 cell cycle: effects of nucleolar protein Bop1 on G1/S tran- protein family,” Blood, vol. 112, no. 13, pp. 5228–5237, 2008. sition,” Molecular and Cellular Biology, vol. 21, no. 13, pp. [84] N. C. Jones, M. L. Lynn, K. Gaudenz et al., “Prevention 4246–4255, 2001. of the neurocristopathy Treacher Collins syndrome through [101] C. Deisenroth and Y. Zhang, “Ribosome biogenesis surveil- inhibition of p53 function,” Nature Medicine, vol. 14, no. 2, lance: probing the ribosomal protein-Mdm2-p53 pathway,” pp. 125–133, 2008. Oncogene, vol. 29, no. 30, pp. 4253–4260, 2010. [85] K. A. McGowan, J. Z. Li, C. Y. Park et al., “Ribosomal [102] D. M. Gilkes, L. Chen, and J. Chen, “MDMX regulation of mutations cause p53-mediated dark skin and pleiotropic p53 response to ribosomal stress,” EMBO Journal, vol. 25, no. effects,” Nature Genetics, vol. 40, no. 8, pp. 963–970, 2008. 23, pp. 5614–5625, 2006. [86] M. Barkic,´ S. Crnomarkovic,´ K. Grabusiˇ c´ et al., “The p53 [103] X. X. Sun, Y. G. Wang, D. P. Xirodimas, and M. S. Dai, tumor suppressor causes congenital malformations in Rpl24- “Perturbation of 60 S ribosomal biogenesis results in riboso- deficient mice and promotes their survival,” Molecular and mal protein L5- and L11-dependent p53 activation,” Journal Cellular Biology, vol. 29, no. 10, pp. 2489–2504, 2009. of Biological Chemistry, vol. 285, no. 33, pp. 25812–25821, [87] C. Constantinou, A. Elia, and M. J. Clemens, “Activation of 2010. p53 stimulates proteasome-dependent truncation of elF4E- [104] A. Pellagatti, T. Marafioti, J. C. Paterson et al., “Induction of binding protein 1 (4E-BP1),” Biology of the Cell, vol. 100, no. p53 and up-regulation of the p53 pathway in the human 5q- 5, pp. 279–289, 2008. syndrome,” Blood, vol. 115, no. 13, pp. 2721–2723, 2010. 12 Leukemia Research and Treatment

[105] M. E. Perry, “The regulation of the p53-mediated stress [121] G. Cazzaniga, M. G. Dell’Oro, C. Mecucci et al., “Nucleopho- response by MDM2 and MDM4,” Cold Spring Harbor Per- smin mutations in childhood acute myelogenous leukemia spectives in Biology, vol. 2, no. 1, article a000968, 2010. with normal karyotype,” Blood, vol. 106, no. 4, pp. 1419– [106] J. S. L. Ho, W. Ma, D. Y. L. Mao, and S. Benchimol, “p53- 1422, 2005. dependent transcriptional repression of c-myc is required for [122] R. Rau and P.Brown, “Nucleophosmin (NPM1) mutations in G1cellcyclearrest,”Molecular and Cellular Biology, vol. 25, adult and childhood acute myeloid leukaemia: towards def- no. 17, pp. 7423–7431, 2005. inition of a new leukaemia entity,” Hematological Oncology, [107] M. Sachdeva, S. Zhu, F. Wu et al., “p53 represses c-Myc vol. 27, no. 4, pp. 171–181, 2009. through induction of the tumor suppressor miR-145,” Pro- [123] M. J. Walter, “Del(5q): gene dosage matters,” Blood, vol. 110, ceedings of the National Academy of Sciences of the United no. 2, pp. 473–474, 2007. States of America, vol. 106, no. 9, pp. 3207–3212, 2009. [124] R. Neuwirtova, O. Fuchs, D. Provaznikova et al., “Fli-1 and [108] M. Sachdeva and Y. Y. Mo, “p53 and c-myc: how does the cell EKLF gene expression in patients with MDS 5q- syndrome,” balance ”yin” and ”yang”?” Cell Cycle, vol. 8, no. 9, p. 1303, Blood, vol. 114, pp. 1090–1091, 2009, abstract no. 2788, 2009. Proceedings of the 51st Annual Meeting of the American [109] A. H. L. Truong, D. Cervi, J. Lee, and Y. Ben-David, “Direct Society of Hematology, December 5–8, 2009, New Orleans, transcriptional regulation of MDM2 by Fli-1,” Oncogene, vol. La, USA. 24, no. 6, pp. 962–969, 2005. [125] R. Neuwirtova, O. Fuchs, A. Jonasova et al., “The role of Fli-1 [110] D. H. Christiansen, M. K. Andersen, and J. Pedersen-Bjer- and EKLF gene expression in 5q- syndrome compared to gaard, “Mutations with loss of heterozygosity of p53 are com- MDS low risk with normal chromosome 5,” in Proceedings mon in therapy-related myelodysplasia and acute myeloid of the XXXIII World Congress of the International Society leukemia after exposure to alkylating agents and significantly of Hematology, Jerusalem, Israel, abstract no. 114, October associated with deletion or loss of 5q, a complex karyotype, 2010. and a poor prognosis,” Journal of Clinical Oncology, vol. 19, [126] Y. Ben-David, E. B. Giddens, K. Letwin, and A. Bernstein, no. 5, pp. 1405–1413, 2001. “Erythroleukemia induction by Friend murine leukemia [111] J. Pedersen-Bjergaard, M. K. Andersen, M. T. Andersen, and virus: insertional activation of a new member of the ets gene D. H. Christiansen, “Genetics of therapy-related myelodys- family, Fli-1, closely linked to c-ets-1,” Genes and Develop- plasia and acute myeloid leukemia,” Leukemia, vol. 22, no. 2, ment, vol. 5, no. 6, pp. 908–918, 1991. pp. 240–248, 2008. [127] D. K. Watson, F. E. Smyth, D. M. Thompson et al., “The [112] M. Jadersten,¨ L. Saft, A. Pellagatti et al., “Clonal heterogene- ERGB/Fli-1 gene: isolation and characterization of a new ity in the 5q- syndrome: P53 expressing progenitors pre- member of the family of human ETS transcription factors,” vail during lenalidomide treatment and expand at disease Cell Growth and Differentiation, vol. 3, no. 10, pp. 705–713, progression,” Haematologica, vol. 94, no. 12, pp. 1762–1766, 1992. 2009. [128] D. D. K. Prasad, V. N. Rao, and E. S. P. Reddy, “Structure and [113] M. Jadersten,¨ L. Saft, A. Smith et al., “TP53 mutations in low- expression of human Fli-1 gene,” Cancer Research, vol. 52, no. risk myelodysplastic syndromes with del(5q) predict dinase 20, pp. 5833–5837, 1992. progression,” Journal of Clinical Oncology, vol. 29, pp. 1971– [129] L. Selleri, M. Giovannini, A. Romo et al., “Cloning of the 1979, 2011. entire FLI1 gene, disrupted by the Ewing’s sarcoma transloca- [114] C. M. Lim, M. A. Cater, J. F. B. Mercer, and S. La Fontaine, tion breakpoint on 11q24, in a yeast artificial chromosome,” “Copper-dependent interaction of dynactin subunit p62 with Cytogenetics and Cell Genetics, vol. 67, no. 2, pp. 129–136, the N terminus of ATP7B but not ATP7A,” Journal of Bio- 1994. logical Chemistry, vol. 281, no. 20, pp. 14006–14014, 2006. [130] J. Starck, N. Cohet, C. Gonnet et al., “Functional cross- [115] S. R. Patel, J. L. Richardson, H. Schulze et al., “Differential antagonism between transcription factors FLI-1 and EKLF,” roles of microtubule assembly and sliding in proplatelet Molecular and Cellular Biology, vol. 23, no. 4, pp. 1390–1402, formation by megakaryocytes,” Blood, vol. 106, no. 13, pp. 2003. 4076–4085, 2005. [131] M. Eisbacher, M. L. Holmes, A. Newton et al., “Protein- [116] J. E. Italiano Jr., S. Patel-Hett, and J. H. Hartwig, “Mechan- protein interaction between Fli-1 and GATA-1 mediates syn- ics of proplatelet elaboration,” Journal of Thrombosis and ergistic expression of megakaryocyte-specific genes through Haemostasis, vol. 5, supplement 1, pp. 18–23, 2007. cooperative DNA binding,” Molecular and Cellular Biology, [117] F. He, C. T. Wang, and L. T. Gou, “RNA-binding motif pro- vol. 23, no. 10, pp. 3427–3441, 2003. tein RBM22 is required for normal development of zebrafish [132] P. Jackers, G. Szalai, O. Moussa, and D. K. Watson, “Ets- embryos,” Genetics and Molecular Research,vol.8,no.4,pp. dependent regulation of target gene expression during meg- 1466–1473, 2009. akaryopoiesis,” Journal of Biological Chemistry, vol. 279, no. [118] S. Grisendi, R. Bernardi, M. Rossi et al., “Role of nucleophos- 50, pp. 52183–52190, 2004. min in embryonic development and tumorigenesis,” Nature, [133] J. L. Svenson, K. Chike-Harris, M. Y. Amria, and T. K. vol. 437, no. 7055, pp. 147–153, 2005. Nowling, “The mouse and human Fli1 genes are similarly [119] S. Grisendi, C. Mecucci, B. Falini, and P. P. Pandolfi, “Nucle- regulated by Ets factors in T cells,” Genes and Immunity, vol. ophosmin and cancer,” Nature Reviews Cancer, vol. 6, no. 7, 11, no. 2, pp. 161–172, 2010. pp. 493–505, 2006. [134] J. Starck, A. Doubeikovski, S. Sarrazin et al., “Spi-1/PU.1 Is a [120] P. Sportoletti, S. Grisendi, S. M. Majid et al., “Npm1 is a hap- positive regulator of the Fli-1 gene involved in inhibition of loinsufficient suppressor of myeloid and lymphoid malig- erythroid differentiation in friend erythroleukemic cell lines,” nancies in the mouse,” Blood, vol. 111, no. 7, pp. 3859–3862, Molecular and Cellular Biology, vol. 19, no. 1, pp. 121–135, 2008. 1999. Leukemia Research and Treatment 13

[135] N. Rekhtman, F. Radparvar, T. Evans, and A. I. Skoultchi, [152] Y. Liu, S. E. Elf, Y. Miyata et al., “p53 regulates hematopoietic “Direct interaction of hematopoietic transcription factors stem cell quiescence,” Cell Stem Cell, vol. 4, no. 1, pp. 37–48, PU.1 and GATA- 1: functional antagonism in erythroid cells,” 2009. Genes and Development, vol. 13, no. 11, pp. 1398–1411, 1999. [153] Y. Liu, S. E. Elf, T. Asai et al., “The p53 tumor suppressor [136] P. Zhang, X. Zhang, A. Iwama et al., “PU.1 inhibits GATA-1 protein is a critical regulator of hematopoietic stem cell function and erythroid differentiation by blocking GATA-1 behavior,” Cell Cycle, vol. 8, no. 19, pp. 3120–3124, 2009. DNA binding,” Blood, vol. 96, no. 8, pp. 2641–2648, 2000. [154] E. Wattel, C. Preudhomme, B. Hecquet et al., “p53 Mutations [137] G. Juban, G. Giraud, B. Guyot et al., “Spi-1 and Fli-1 di- are associated with resistance to chemotherapy and short rectly activate common target genes involved in ribosome survival in hematologic malignancies,” Blood, vol. 84, no. 9, biogenesis in friend erythroleukemic cells,” Molecular and pp. 3148–3157, 1994. Cellular Biology, vol. 29, no. 10, pp. 2852–2864, 2009. [155] Y. Kita-Sasai, S. Horiike, S. Misawa et al., “International [138] D. T. Starczynowski and A. Karsan, “Innate immune signal- prognostic scoring system and TP53 mutations are indepen- ing in the myelodysplastic syndromes,” Hematology/Oncology dent prognostic indicators for patients with myelodysplastic Clinics of North America, vol. 24, pp. 343–359, 2010. syndrome,” British Journal of Haematology, vol. 115, no. 2, pp. 309–312, 2001. [139] D. R. Hodge, W. Xiao, P. A. Clausen, G. Heidecker, M. Szyf, [156] S. Horiike, Y. Kita-Sasai, M. Nakao, and M. Taniwaki, “Con- and W. L. Farrar, “Interleukin-6 regulation of the human figuration of the TP53 gene as an independent prognostic DNA methyltransferase (HDNMT) gene in human ery- parameter of myelodysplastic syndrome,” Leukemia and throleukemia cells,” Journal of Biological Chemistry, vol. 276, Lymphoma, vol. 44, no. 6, pp. 915–922, 2003. no. 43, pp. 39508–39511, 2001. [157] L. Garderet, L. Kobari, C. Mazurier et al., “Unimpaired [140] D. R. Hodge, D. Li, S. M. Qi, and W. L. Farrar, “IL-6 terminal erythroid differentiation and preserved enucleation induces expression of the Fli-1 proto-oncogene via STAT3,” capacity in myelodysplastic 5q(del) clones: a single cell Biochemical and Biophysical Research Communications, vol. study,” Haematologica, vol. 95, no. 3, pp. 398–405, 2010. 292, pp. 287–291, 2002. [158] T. M. A. Neildez-Nguyen, H. Wajcman, M. C. Marden et al., [141] M. R. Tallack, T. Whitington, W. S. Yuen et al., “A global role “Human erythroid cells produced ex vivo at large scale dif- for KLF1 in erythropoiesis revealed by ChIP-seq in primary ferentiate into red blood cells in vivo,” Nature Biotechnology, erythroid cells,” Genome Research, vol. 20, no. 8, pp. 1052– vol. 20, no. 5, pp. 467–472, 2002. 1063, 2010. [159] M. C. Giarratana, L. Kobari, H. Lapillonne et al., “Ex vivo [142] M. Siatecka and J. J. Bieker, “The multifunctional role generation of fully mature human red blood cells from hema- of EKLF/KLF1 during erythropoiesis,” Blood, vol. 118, pp. topoietic stem cells,” Nature Biotechnology, vol. 23, no. 1, pp. 2044–2054, 2011. 69–74, 2005. [143] L. C. Dore´ and J. D. Crispino, “Transcription factor in [160] M. Jadersten,¨ “Pathophysiology and treatment of the myelo- erythroid cell and megakaryocyte development,” Blood, vol. dysplastic syndrome with isolated 5q deletion,” Haematolog- 118, pp. 231–239, 2011. ica, vol. 95, no. 3, pp. 348–351, 2010. [144] J. Borg, P. Papadopoulos, M. Georgitsi et al., “Haploinsuf- [161] M. Virgilio, E. Payne, A. Narla et al., “Treatment of zebra- ficiency for the erythroid transcription factor KLF1 causes fish models of ribosomopathies (Diamond Blackfan ane- hereditary persistence of fetal hemoglobin,” Nature Genetics, mia (DBA) and 5q-syndrome with Lleucine results in an vol. 42, no. 9, pp. 801–805, 2010. improvement of anemia and development defects: evidence [145] P. Frontelo, D. Manwani, M. Galdass et al., “Novel role for for a common pathway,” Blood, vol. 116, abstract 195, pp. 89– EKLF in megakaryocyte lineage commitment,” Blood, vol. 90, 2010. 110, no. 12, pp. 3871–3880, 2007. [162] J. Cmejlova, L. Dolezalova, D. Pospisilova, K. Petrtylova, J. ffi [146] F. Bouilloux, G. Juban, N. Cohet et al., “EKLF restricts Petrak, and R. Cmejla, “Translational e ciency in patients megakaryocytic differentiation at the benefit of erythrocytic with Diamond-Blackfan anemia,” Haematologica, vol. 91, no. differentiation,” Blood, vol. 112, no. 3, pp. 576–584, 2008. 11, pp. 1456–1464, 2006. [163] J. C. Anthony, T. G. Anthony, S. R. Kimball, T. C. Vary, and [147] O. Klimchenko, M. Mori, A. DiStefano et al., “A common L. S. Jefferson, “Orally administered leucine stimulates pro- bipotent progenitor generates the erythroid and megakary- tein synthesis in skeletal muscle of postabsorptive rats ocyte lineages in embryonic stem cell-derived primitive in association with increased elF4F formation,” Journal of hematopoiesis,” Blood, vol. 114, no. 8, pp. 1506–1517, 2009. Nutrition, vol. 130, no. 2, pp. 139–145, 2000. [148] M. R. Tallack and A. C. Perkins, “Megakaryocyte-erythroid [164] C. J. Lynch, B. J. Patson, J. Anthony et al., “Leucine is a lineage promiscuity in EKLF null mouse blood,” Haemato- direct-acting nutrient signal that regulates protein synthesis logica, vol. 95, no. 1, pp. 144–147, 2010. in adipose tissue,” American Journal of Physiology, vol. 283, ffi [149] S. Dutt, A. Narla, K. Lin et al., “Haploinsu ciency for pp. E506–E513, 2002. ribosomal protein genes causes selective activation of p53 in [165] L. E. Norton and D. K. Layman, “Leucine regulates trans- human erythroid progenitor cells,” Blood, vol. 117, no. 9, pp. lation initiation of protein synthesis in skeletal muscle after 2567–2576, 2011. excercise,” Journal of Nutrition, vol. 136, pp. 533S–537S, 2006. [150] M. Cazzola, “Myelodysplastic syndrome with isolated 5q [166] J. Escobar, J. W. Frank, A. Suryawan et al., “Amino acid deletion (5q- syndrome). A clonal stem cell disorder char- availability and age affect the leucine stimulation of protein acterized by defective ribosome biogenesis,” Haematologica, synthesis and eIF4F formation in muscle,” American Journal vol. 93, no. 7, pp. 967–972, 2008. of Physiology, vol. 293, pp. E1615–E1621, 2007. [151] J. L. Barlow, L. F. Drynan, N. L. Trim, W. N. Erber, A. [167] D. Pospisilova, J. Cmejlova, J. Hak, T. Adam, and R. Cmejla, J. Warren, and A. N. J. McKenzie, “New insights into 5q- “Successful treatment of a Diamond-Blackfan anemia patient syndrome as a ribosomopathy,” Cell Cycle, vol. 9, no. 21, pp. with amino acid leucine,” Haematologica,vol.92,no.5,pp. 4286–4293, 2010. e66–67, 2007. 14 Leukemia Research and Treatment

[168] A. List, G. Dewald, J. Bennett et al., “Lenalidomide in the patients with myelodysplastic syndrome (MDS) after in vitro myelodysplastic syndrome with chromosome 5q deletion,” exposure to lenalidomide correlates with clinical response in The New England Journal of Medicine, vol. 355, no. 14, pp. patients harboring the del5q abnormality,” Blood, vol. 116, 1456–1465, 2006. abstract 123, p. 60, 2010. [169] M. Melchert, V. Kale, and A. List, “The role of lenalidomide [184] M. Ximeri, A. Galanopoulos, M. Klaus et al., “Effect of lenal- in the treatment of patients with chromosome 5q deletion idomide therapy on hematopoiesis of patients with myelo- and other myelodysplastic syndromes,” Current Opinion in dysplastic syndrome associated with chromosome 5q dele- Hematology, vol. 14, no. 2, pp. 123–129, 2007. tion,” Haematologica, vol. 95, no. 3, pp. 406–414, 2010. [170] A. F. List, “Lenalidomide—the phoenix rises,” The New [185] R. Tehranchi, P. S. Woll, K. Anderson et al., “Persistent malig- England Journal of Medicine, vol. 357, no. 21, pp. 2183–2186, nant stem cells in del(5q) myelodysplasia in remission,” The 2007. New England Journal of Medicine, vol. 363, no. 11, pp. 1025– [171] S. Kurtin and A. List, “Durable long-term responses in 1037, 2010. patients with myelodysplastic syndromes treated with len- alidomide,” Clinical Lymphoma and Myeloma,vol.9,no.3, pp. E10–E13, 2009. [172] R. S. Komrojki and A. F. List, “Lenalidomide for teatment of myelodysplastic syndromes: current status and future direc- tions,” Hematology/Oncology Clinics of North America, vol. 24, pp. 377–388, 2010. [173] S. M. Post and A. Quintas-Cardama,´ “Closing in on the pathogenesis of the 5q- syndrome,” Expert Review of Anti- cancer Therapy, vol. 10, no. 5, pp. 655–658, 2010. [174] A. Raza, J. A. Reeves, E. J. Feldman et al., “Phase 2 study of lenalidomide in transfusion-dependent, low-risk, and intermediate-1-risk myelodysplastic syndromes with kary- otypes other than deletion 5q,” Blood, vol. 111, no. 1, pp. 86– 93, 2008. [175] B. L. Ebert, N. Galili, P. Tamayo et al., “An erythroid differentiation signature predicts response to lenalidomide in myelodysplastic syndrome,” PLoS Medicine, vol. 5, no. 2, pp. 0312–0322, 2008. [176] C. Chen, D. T. Bowen, A. A. N. Giagounidis, B. Schlegel- berger, S. Haase, and E. G. Wright, “Identification of dis- ease- and therapy-associated proteome changes in the sera of patients with myelodysplastic syndromes and del(5q),” Leukemia, vol. 24, no. 11, pp. 1875–1884, 2010. [177] S. Wei, X. Chen, K. Rocha et al., “A critical role for phos- phatase haplodeficiency in the selective suppression of dele- tion 5q MDS by lenalidomide,” Proceedings of the National Academy of Sciences of the United States of America, vol. 106, no. 31, pp. 12974–12979, 2009. [178] S. H. Kimura and H. Nojima, “Cyclin G1 associates with MDM2 and regulates accumulation and degradation of p53 protein,” Genes to Cells, vol. 7, no. 8, pp. 869–880, 2002. [179] X. K. Zhang and D. K. Watson, “The FLI-1 transcription factor is a short-lived phosphoprotein in T cells,” Journal of Biochemistry, vol. 137, no. 3, pp. 297–302, 2005. [180] J. B. Bartlett, K. Dredge, and A. G. Dalgleish, “The evolution of thalidomide and its IMiD derivatives as anticancer agents,” Nature Reviews Cancer, vol. 4, no. 4, pp. 314–322, 2004. [181] A. Pellagatti, M. Jadersten,¨ A. M. Forsblom et al., “Lenalido- mide inhibits the malignant clone and up-regulates the SPARC gene mapping to the commonly deleted region in 5q- syndrome patients,” Proceedings of the National Academy of Sciences of the United States of America, vol. 104, no. 27, pp. 11406–11411, 2007. [182] E. N. Oliva, M. Cuzzola, F. Nobile et al., “Changes in RPS14 expression levels during lenalidomide treatment in Low- and Intermediate-1-risk myelodysplastic syndromes with chro- mosome 5q deletion,” European Journal of Haematology, vol. 85, no. 3, pp. 231–235, 2010. [183] C. P. Venner, A. F. List, T. J. Nevill et al., “Induction of micro RNA-143 and 145 in pre-treatment CD34+ cells from